EUVIMEDEUVIMED entdecken
Crossref

TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning

Yuxuan Li, Yicheng Gao, Ning Yang, Stephen Xia

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (28) · 23301-23309

DOI: 10.1609/aaai.v40i28.39499 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yuxuan Li und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Episodic tasks in Reinforcement Learning (RL) often pose challenges due to sparse reward signals and high-dimensional state spaces, which hinder efficient learning. Additionally, these tasks often feature hidden “trap states”—irreversible failures that prevent task completion but do not provide explicit negative rewards to guide agents away from repeated errors. To address these issues, we propose Time-Weighted Contrastive Reward Learning (TW-CRL), an Inverse Reinforcement Learning (IRL) framework that leverages both successful and failed demonstrations. By incorporating temporal information, TW-CRL learns a dense reward function that identifies critical states associated with success or failure. This approach not only enables agents to avoid trap states but also encourages meaningful exploration beyond simple imitation of expert trajectories. Empirical evaluations on navigation tasks and robotic manipulation benchmarks demonstrate that TW-CRL surpasses state-of-the-art methods, achieving improved efficiency and robustness.

Zitieren

APA

Li, Y., Gao, Y., Yang, N., & Xia, S. (2026). TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(28), 23301-23309. https://doi.org/10.1609/aaai.v40i28.39499

Vancouver

Li Y, Gao Y, Yang N, Xia S. TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(28):23301-23309. doi:10.1609/aaai.v40i28.39499

BibTeX
@article{Li2026,
  author = {Li, Yuxuan and Gao, Yicheng and Yang, Ning and Xia, Stephen},
  title = {TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {28},
  pages = {23301--23309},
  doi = {10.1609/aaai.v40i28.39499}
}
Kostenlos starten →